Users' documentation for BLink in c++


Installation

Compiled version:

Compiled version for linux.-After you have succesfully obtained the archive you can execute the following Unix commands to install the program:

Source code version:

Source code version for linux.-After you have succesfully obtained the archive you can execute the following Unix commands to install the program:

Execution

To use BLink, use the following command line:

./BLink <filename> [<isPhased (0: no default, 1: yes)>] [<useBayesianMeasure (1: yes default, 0:no)>]

where <filename> is the input file with format as shown below.

By default, input data are considered unphased and the Bayesian method is used. Use 1 as a second argument is the phase is solved (alleles are ordered). Use 0 as a third argument if the original (MLE) LD measures are to be computed.

 

Format of the input genotype file

A text file of SNP data set, in post-makeped format as described here:

One row per individual, 7 + 2xTotalSNPs columns, with TotalSNPs being the number of SNPs. Columns must be separared by a blank or tab character.Each column must contain an integer. See an example of input file.

Description of columns:

Format of the input file with physical positions

One position per row. Click here to see an example.

Output for the program BLink

The output will be written at a file with the same name as the input file but with the extension ".pm". The output file is a csv text file with one line for each pair of SNPs with the number and positions of each snp, haplotype and allele frequencies and the Bayesian estimation of D, D', r2 and Yules Q.